An automatic speech recognition (ASR) system determines a semantic meaning of a speech input. Typically, the input speech is processed into a sequence of digital speech feature frames. Each speech feature frame can be thought of as a multi-dimensional vector that represents various characteristics of the speech signal present during a short time window of the speech. For example, the multi-dimensional vector of each speech frame can be derived from cepstral features of the short time Fourier transform spectrum of the speech signal (MFCCs)—the short time power or component of a given frequency band—as well as the corresponding first- and second-order derivatives (“deltas” and “delta-deltas”). In a continuous recognition system, variable numbers of speech frames are organized as “utterances” representing a period of speech followed by a pause, which in real life loosely corresponds to a spoken sentence or phrase.
The ASR system compares the input utterances to find statistical acoustic models that best match the vector sequence characteristics and determines corresponding representative text associated with the acoustic models. More formally, given some input observations A, the probability that some string of words W were spoken is represented as P(W|A), where the ASR system attempts to determine the most likely word string:
      W    ^    =                    arg        ⁢                                  ⁢        max            W        ⁢                  ⁢          P      ⁡              (                  W          ❘          A                )            Given a system of statistical acoustic models, this formula can be re-expressed as:
      W    ^    =                    arg        ⁢                                  ⁢        max            W        ⁢                  ⁢          P      ⁡              (        W        )              ⁢          P      ⁡              (                  A          ❘          W                )            where P(A|W) corresponds to the acoustic models and P(W) reflects the prior probability of the word sequence as provided by a statistical language model reflecting the probability of given word in the recognition vocabulary occurring.
The acoustic models are typically probabilistic state sequence models such as hidden Markov models (HMMs) that model speech sounds using mixtures of probability distribution functions (Gaussians). Acoustic models often represent phonemes in specific contexts, referred to as PELs (Phonetic Elements), e.g. triphones or phonemes with known left and/or right contexts. State sequence models can be scaled up to represent words as connected sequences of acoustically modeled phonemes, and phrases or sentences as connected sequences of words. When the models are organized together as words, phrases, and sentences, additional language-related information is also typically incorporated into the models in the form of a statistical language model.
The words or phrases associated with the best matching model structures are referred to as recognition candidates or hypotheses. A system may produce a single best recognition candidate—the recognition result—or multiple recognition hypotheses in various forms such as an N-best list, a recognition lattice, or a confusion network. Further details regarding continuous speech recognition are provided in U.S. Pat. No. 5,794,189, entitled “Continuous Speech Recognition,” and U.S. Pat. No. 6,167,377, entitled “Speech Recognition Language Models,” the contents of which are incorporated herein by reference.
One specific application of ASR technology is for automatic transcription of real-world audio from speech sessions with multiple different speakers such as teleconferences, meeting records, police interviews, etc. There appears to be a large commercial market for accurate but inexpensive automatic transcription of such speech sessions. In many specific contexts the combination of the acoustic conditions and the speaking styles mean that current state-of-the-art ASR technology is unable to provide an accurate transcription. Manual transcriptions by human transcriptionists are generally very slow and expensive to obtain because of the time required—human transcriptionists typically need to listen to the audio many times.